After reading about light polarization I understood, that if light is polarized:
- circularly left then the spin of each photon is parallel to the velocity
- circularly right then the spin of each photon is antiparallel to the velocity
- elliptically - then there's more one photons, then the others
- linearly - then there's exactly the same number of both spins.
But now my question is - if my above understanding is correct, then how can be linearly polarized light different, than not polarized at all? Both have the same amount of photons having spin parallel and antiparallel to the velocity.
In other words - what is the relation between spin of each photon (or distribution of spins of all photons) and polarization plane?
The problem can be rephrased as follows: when you consider a single photon, where does it "know" from whether it can go through a polarizer or no? Is there some vector property of a single photon, that must be aligned with the polarization plane of the filter? What is this property?